python中的字符串解析/替换（使用reg表达式）

我正在尝试编写一个解析器：将字符串转换为查询格式。并卡在字符串替换的特定点（通过匹配模式）

我无法找出正则表达式模式匹配

我有一个输入字符串，比如

ip_query_string = "CITY == 'Mumbai' & LOCATION in ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"

#Mark the """& after CITY == """ and  """ in after LOCATION""".
#Then there is another "& and a string ' in '" inside values for in-condition.

#My output should be:
op_query_string = "CITY == 'Mumbai' AND LOCATION IN ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"

#if i will find  ' & ' or a ' in ' (before and after there are spaces):: I have to replace them with ' AND ' and ' IN ' respectively.(In this case a ip_string.replace(' & ', ' AND ').replace(' in ', ' In ')) would work.BUT read next point.
#And if they are inside a in-condition values like 'Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai' then don't replace them. keep them as is.
#If you look at op_string in-condition, the & and in are not replaced.

请帮忙形成一个逻辑

---

或者reg模式是什么（if&或在单引号中加上其他字符，不要替换，否则替换）？

以某种奇怪的方式工作（可能不是pythonic），但它工作了

def rplc_str(s):
   sp = s.split("'")
   print('After split==',sp)
   sp1 = [x.replace(' & ', ' AND ') if ((x.startswith(' &')) or (x.startswith('] &'))) else x for x in sp]
   print('After replacing & ==',sp1)
   sp2 = [x.replace(' in ', ' IN ') if x.endswith(' [') else x for x in sp1]
   print('After replacing in ==',sp1)
   return "'".join(sp2)

ip_str = "CITY == 'Mumbai' & LOCATION in ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"
op_str = rplc_str(ip_str)
print(op_str)
#After split== ['CITY == ', 'Mumbai', ' & LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#After replace & == ['CITY == ', 'Mumbai', ' AND LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#After replace in == ['CITY == ', 'Mumbai', ' AND LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#CITY == 'Mumbai' AND LOCATION IN ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']

---

希望它能帮助某些人，但仍在等待一些类似于python的答案（我的意思是reg expr.）

以某种奇怪的方式实现了它（可能不是python），但它成功了

def rplc_str(s):
   sp = s.split("'")
   print('After split==',sp)
   sp1 = [x.replace(' & ', ' AND ') if ((x.startswith(' &')) or (x.startswith('] &'))) else x for x in sp]
   print('After replacing & ==',sp1)
   sp2 = [x.replace(' in ', ' IN ') if x.endswith(' [') else x for x in sp1]
   print('After replacing in ==',sp1)
   return "'".join(sp2)

ip_str = "CITY == 'Mumbai' & LOCATION in ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"
op_str = rplc_str(ip_str)
print(op_str)
#After split== ['CITY == ', 'Mumbai', ' & LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#After replace & == ['CITY == ', 'Mumbai', ' AND LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#After replace in == ['CITY == ', 'Mumbai', ' AND LOCATION in [', 'Harrys Bar & Cafe: Mumbai', ',', 'Hard Rock Cafe in Mumbai', ']']
#CITY == 'Mumbai' AND LOCATION IN ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']

希望它能帮助一些人，但仍在等待一些pythonic的答案（我的意思是reg expr.）

使用函数的简短解决方案：

import re

ip_query_string = "CITY == 'Mumbai' & LOCATION in ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"
op_query_string  = re.sub(r'^([^[]+?)(in)', r'\1IN', re.sub(r'^([^[]+?)(&)', r'\1AND', ip_query_string))

print(op_query_string)

输出：

CITY == 'Mumbai' AND LOCATION IN ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']

使用函数的简短解决方案：

import re

ip_query_string = "CITY == 'Mumbai' & LOCATION in ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']"
op_query_string  = re.sub(r'^([^[]+?)(in)', r'\1IN', re.sub(r'^([^[]+?)(&)', r'\1AND', ip_query_string))

print(op_query_string)

输出：

CITY == 'Mumbai' AND LOCATION IN ['Harrys Bar & Cafe: Mumbai','Hard Rock Cafe in Mumbai']

---

为什么

'Harry's

应该变成

'Harrys

？对不起，有个打字错误，总是Harrys。编辑了这个问题。为什么哈利的应该变成哈雷的？对不起，有个打字错误，总是哈雷的。在本例中编辑了question.works（问题），但如果在正常条件的值中有“&”或“in”，则编辑失败。比如ip_query_string=“CITY==‘果阿的孟买’、‘孟买的硬石咖啡馆’、‘孟买的哈雷酒吧和咖啡馆’”……对于城市价值，我在其中加了一个“in”。然后，上述reg_expr结果为“城市==‘果阿孟买’和地点[‘孟买Harrys酒吧和咖啡馆’，‘孟买硬石咖啡馆’”…即，将‘果阿孟买’改为‘果阿孟买’。这是不应该发生的。如果在单引号中发现“&”和“in”，则不应更改。在这种情况下有效（问题），但如果在正常条件的值中有“&”或“in”，则失败。比如ip_query_string=“CITY==‘果阿的孟买’、‘孟买的硬石咖啡馆’、‘孟买的哈雷酒吧和咖啡馆’”……对于城市价值，我在其中加了一个“in”。然后，上述reg_expr结果为“城市==‘果阿孟买’和地点[‘孟买Harrys酒吧和咖啡馆’，‘孟买硬石咖啡馆’”…即，将‘果阿孟买’改为‘果阿孟买’。这是不应该发生的。如果在单引号内发现“&”和“in”，则不应更改。